Three-dimensional ultrasound diagnosis of adenomyosis is not associated with adverse pregnancy outcome following single thawed euploid blastocyst transfer: prospective cohort study.

OBJECTIVES: The objectives of this study were (1) to assess the prevalence of ultrasound (US) features of adenomyosis in an infertile population undergoing in-vitro fertilization (IVF), (2) to define the inter- and intrarater agreement of three-dimensional (3D) US assessment of adenomyosis, and (3) to evaluate sonographic features of adenomyosis with respect to pregnancy outcome following transfer of a single thawed euploid blastocyst.

METHODS: This was a prospective cohort study. Subjects scheduled to undergo a single thawed euploid blastocyst transfer between April and December 2017 at a large IVF center were eligible for inclusion. Enrolled subjects underwent endometrial preparation for frozen embryo transfer. 3D-US was performed on the day prior to embryo transfer, with images stored for subsequent evaluation. Subjects then underwent transfer of a single thawed euploid blastocyst, and pregnancy outcomes were collected. All 3D-US volumes were de-identified and reviewed independently by five reproductive endocrinologists/infertility specialists with expertise in gynecological US for the presence of seven sonographic features of adenomyosis: global uterine enlargement, myometrial wall asymmetry, heterogeneous echogenicity, irregular junctional zone, myometrial cysts, fan-shaped shadowing and ill-defined myometrial lesions. Adenomyosis was considered to be present if the majority of the reviewers noted at least one of the seven sonographic features. Inter- and intrarater agreement was evaluated using Fleiss's kappa. Clinical and cycle characteristics of subjects with and those without adenomyosis were compared. The primary outcome of interest was live birth rate. Secondary outcomes included clinical pregnancy rate and miscarriage rate. Logistic regression analysis was performed to account for potential confounders.

RESULTS: A total of 648 subjects were included. The prevalence of adenomyosis on US was 15.3% (99/648). On retrospective chart review, very few patients with adenomyosis had symptoms. The inter- and intrarater agreement amongst five independent specialists conducting the 3D-US assessments of adenomyosis were poor (κ = 0.23) and moderate (κ = 0.58), respectively. Subjects with adenomyosis were older (37.1 vs 35.9 years, P = 0.02) and more likely to undergo a gonadotropin-releasing hormone agonist downregulation protocol when compared with those without adenomyosis (12.1% vs 5.1%, P = 0.02). Clinical pregnancy (80.0% vs 75.0%) and live birth (69.5% vs 66.5%) rates were similar between the groups. When adjusting for potential confounders, there was no difference in the rate of clinical pregnancy (adjusted odds ratio (aOR), 1.47 (95% CI, 0.85-2.56)), miscarriage (aOR, 1.3 (95% CI, 0.62-2.72)) or live birth (aOR, 1.28 (95% CI, 0.78-2.08)) between subjects with and those without adenomyosis. No individual sonographic marker of adenomyosis was predictive of pregnancy outcome.

CONCLUSIONS: The inter-rater agreement of 3D-US assessment of adenomyosis is poor. Furthermore, sonographic markers of adenomyosis in asymptomatic patients may not be associated with altered pregnancy outcome following transfer of a single thawed euploid blastocyst. These findings suggest that routine screening for asymptomatic adenomyosis in an unselected infertile patient population undergoing frozen embryo transfer may not be warranted. © 2020 International Society of Ultrasound in Obstetrics and Gynecology.